DvD: Unleashing a Generative Paradigm for Document Dewarping via Coordinates-based Diffusion Model

TL;DR

This paper introduces DvD, a novel generative diffusion model for document dewarping that uses coordinate-level denoising and a time-variant refinement mechanism, achieving state-of-the-art results and providing a new large-scale benchmark.

Contribution

It presents the first diffusion-based generative model for document dewarping with coordinate-level denoising and a new comprehensive benchmark dataset.

Findings

DvD achieves state-of-the-art performance on multiple benchmarks.

Coordinate-level denoising effectively preserves document structures.

The new benchmark enables more thorough evaluation of dewarping models.

Abstract

Document dewarping aims to rectify deformations in photographic document images, thus improving text readability, which has attracted much attention and made great progress, but it is still challenging to preserve document structures. Given recent advances in diffusion models, it is natural for us to consider their potential applicability to document dewarping. However, it is far from straightforward to adopt diffusion models in document dewarping due to their unfaithful control on highly complex document images (e.g., 2000$times$3000 resolution). In this paper, we propose DvD, the first generative model to tackle document Dewarping via a Diffusion framework. To be specific, DvD introduces a coordinate-level denoising instead of typical pixel-level denoising, generating a mapping for deformation rectification. In addition, we further propose a time-variant condition refinement mechanism to enhance the preservation of document structures. In experiments, we find that current document dewarping benchmarks can not evaluate dewarping models comprehensively. To this end, we present AnyPhotoDoc6300, a rigorously designed large-scale document dewarping benchmark comprising 6,300 real image pairs across three distinct domains, enabling fine-grained evaluation of dewarping models. Comprehensive experiments demonstrate that our proposed DvD can achieve state-of-the-art performance with acceptable computational efficiency on multiple metrics across various benchmarks, including DocUNet, DIR300, and AnyPhotoDoc6300. The new benchmark and code will be publicly available at https://github.com/hanquansanren/DvD.